CN118053280A - Real-time early warning method and system for fire risk of artificial intelligent electric circuit - Google Patents
Real-time early warning method and system for fire risk of artificial intelligent electric circuit Download PDFInfo
- Publication number
- CN118053280A CN118053280A CN202410455501.2A CN202410455501A CN118053280A CN 118053280 A CN118053280 A CN 118053280A CN 202410455501 A CN202410455501 A CN 202410455501A CN 118053280 A CN118053280 A CN 118053280A
- Authority
- CN
- China
- Prior art keywords
- section
- sub
- abnormal
- risk
- subsection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000002159 abnormal effect Effects 0.000 claims description 168
- 238000012544 monitoring process Methods 0.000 claims description 53
- 238000004458 analytical method Methods 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 20
- 238000007405 data analysis Methods 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 claims description 4
- 238000010835 comparative analysis Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 6
- 238000012423 maintenance Methods 0.000 description 3
- 230000004069 differentiation Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Alarm Systems (AREA)
Abstract
The application discloses a real-time early warning method and a real-time early warning system for fire risks of an artificial intelligent electric circuit, and relates to the technical field of early warning for fire risks. The real-time early warning method and system for the fire risk of the artificial intelligent electric circuit provided by the application have the advantages that the efficiency of the early warning work for the fire risk of the circuit is improved.
Description
Technical Field
The application relates to the technical field of fire risk early warning, in particular to an artificial intelligent electric circuit fire risk real-time early warning method and system.
Background
Electrical circuit fires are one of the most common fire types in modern society, and have high occurrence frequency and large loss, and form serious threat to life safety of people. The occurrence of electrical circuit fires is often accompanied by short-circuiting, overload, poor contact, etc. of the circuit, which can cause the temperature of the wires to rise, thereby causing the fire. Therefore, the method has important practical significance in effectively early warning and preventing the fire disaster of the electric circuit.
Traditional electrical circuit fire early warning mainly relies on manual inspection and periodic maintenance, and this kind of mode inefficiency can't realize real-time early warning. Although modern electrical circuit fire early warning technology has made remarkable progress, challenges are faced in how to process a large amount of monitoring data, so as to improve the accuracy and timeliness of real-time early warning of electrical circuits, and further research and exploration are required for solving the problems.
Disclosure of Invention
In order to overcome the defects of the prior art, the application provides an artificial intelligent electric circuit fire risk real-time early warning method and system.
In a first aspect, the application provides a real-time early warning method for fire risk of an artificial intelligent electric circuit, which comprises the following steps:
Acquiring a first target line, performing anomaly on all subsections segmented in the first target line to obtain a negative subsection set, and performing normal sign marking on all subsections segmented in the first target line to obtain a positive subsection set;
Performing independent analysis on the number of the negative sub-section sets and the positions of the negative sub-section sets to obtain a first abnormal sub-section to be judged, and analyzing the positions of the negative sub-section sets to obtain a first-stage risk judgment result when the number of the negative sub-section sets is smaller than the number of the positive sub-section sets;
analyzing the position connection degree of the negative sub-section set to obtain a second-stage risk judging result or a second abnormal sub-section to be judged, and analyzing the comparability of the number of the negative sub-section sets being greater than that of the positive sub-section sets to obtain a first-stage risk judging result;
Acquiring a second target line, carrying out data analysis on a first reference comparison section segmented in the second target line and a first abnormal subsection to be judged, and outputting a third-level risk judgment result;
Performing data analysis on the second reference comparison section set segmented in the second target line and the second abnormal subsection set to be judged, and outputting a primary level risk judgment result;
And based on the primary level risk judging result, carrying out ratio solving processing on the data in the negative subsection set and the positive subsection set, outputting a second level risk judging result, carrying out real-time early warning notification on the first level risk judging result or the second level risk judging result or the third level risk judging result, and outputting a line fire real-time early warning notification result.
Through adopting above-mentioned technical scheme, through carrying out the refinement branch of a plurality of sub-district circuits to first target circuit to the realization carries out the analysis of variety distinguishability with a large amount of data resources, improves the accuracy of data change processing under the multiple condition, and carries out the analysis of the research nature to the change development characteristic of data, strengthens the scientificity of data processing result, carries out the early warning notice of distinguishability according to the conflagration risk decision result under the multiple different conditions of analysis, makes the analysis of line conflagration risk carry out comparatively comprehensive consideration, avoids great error and lower accuracy that the data analysis of uniformity that adopts because of traditional technique caused.
Preferably, a first target line is obtained, the first target line is subjected to segmentation of a monitoring section and a reserved section, and the monitoring section is subjected to refinement of at least two equidistant sections to obtain a first sub-section set;
Collecting a monitoring dataset of all subsections in the first subsection set;
Presetting a monitoring judgment section value data set, carrying out comparative analysis on corresponding section values on the monitoring data set and the monitoring judgment section value data set, and outputting a data judgment normal result and a data judgment abnormal result;
based on the data, judging a normal result, and marking a forward symbol on normal item data to obtain a forward data item set;
Based on the data judging abnormal result, marking the abnormal item data with negative sign to obtain a negative data item set;
Calculating the data item number ratio of the positive data item set to the negative data item set, and preliminarily judging the corresponding sub-section as abnormal when the data item number ratio is smaller than or equal to a preset ratio, marking negative signs to obtain a negative sub-section, and outputting the negative sub-section set according to the negative sub-section;
When the number ratio of the data items is larger than the preset ratio, the corresponding sub-section is primarily judged to be normal, the forward symbol is marked to obtain a forward sub-section, and the forward sub-section set is output according to the forward sub-section.
By adopting the technical scheme, whether the data contained in the plurality of the cut subsections are abnormal or not and the number of abnormal data items appear are marked by two different signs on the corresponding subsections, so that the fire risk is likely to exist on the lines of the subsections in a visual way, the fire risk is analyzed and early-warned in a distinguishing way, and the fire risk level caused by the difference of the positions and the number of the corresponding negative subsections is different, so that the differential analysis is needed.
Preferably, the number and the positions of the negative sub-section sets are judged, when only one negative sub-section set is provided and the negative sub-section is positioned at the junction of the reserved sections, further analysis on whether the risk of line fire exists in the negative sub-section is carried out, and a first abnormal sub-section to be judged is output;
Outputting a first monitoring condition when the number of the negative sub-section sets is more than one and less than the number of the positive sub-section sets, and the distribution positions of the negative sub-section sets are located at the junction of the reserved sections and are continuous among the positions of each negative sub-section;
Judging the line fire risk degree of the first monitoring condition as a first level, and outputting a first-level risk judgment result.
Preferably, when the number of the negative sub-section sets is more than one and less than the number of the positive sub-section sets, and the distribution position of the negative sub-section sets does not fall at the junction of the reserved sections, outputting a second monitoring condition;
Calculating the degree of engagement of the positions of each negative sub-section, and outputting a first degree of engagement;
Presetting a connection degree judging section value, judging the line fire risk degree of a second monitoring condition as a second level according to the first connection degree when the first connection degree falls in the connection degree judging section value, and outputting a second level risk judging result;
When the first degree of engagement does not fall within the degree of engagement judging section value, further analyzing whether the second monitoring condition is in existence of the fire risk of the line, and outputting a second abnormal subsection to be judged;
When the number of the negative subsection sets is greater than that of the positive subsection sets, outputting a third monitoring condition, judging the line fire risk degree of the third monitoring condition as a first level, and outputting a first-level risk judgment result.
By adopting the technical scheme, the diversity analysis under various different conditions is carried out by comparing the quantity contained in the negative subsection set and the quantity contained in the positive subsection set and the difference of the positions of the negative subsection set and the positive subsection set, the misjudgment conditions under which conditions are most likely to occur are extracted at first, and the subsequent further deep analysis processing is carried out, so that the accuracy of the whole data information processing process is greatly improved.
Preferably, a line information base associated with a first target line is obtained, and a second target line with the highest similarity association degree with the first target line is screened out from the line information base;
According to the first sub-section set, the same sub-section of the second target line is segmented to obtain a second sub-section set;
Screening a first reference comparison section corresponding to the first abnormal subsection to be judged from the second subsection set based on the first abnormal subsection to be judged;
acquiring a reference data set of a first reference comparison section, calculating an abnormal difference value between a negative data item set of the first abnormal subsection to be judged and a corresponding data item of the reference data set, and outputting a first abnormal difference value set;
evaluating the mutual influence degree of the data item differences in the first abnormal difference set, and outputting a first influence degree evaluation result;
Presetting an influence degree critical judgment value, and outputting a third-stage risk judgment result when the numerical value in the first influence degree evaluation result is larger than or equal to the influence degree critical judgment value.
By adopting the technical scheme, the second target line with the highest relevance is extracted, the processing steps of the first target line and the second target line are carried out, so that data of the two target lines are compared and subjected to difference processing, whether the first abnormal subsection to be judged is at risk or not and whether fire disaster is found or not are judged according to the finally obtained first abnormal difference value set, further verification and judgment are carried out on the analysis result, and accuracy of the analysis result is enhanced.
Preferably, a second reference comparison section set corresponding to the position of the second abnormal subsection set to be judged is screened out from the second subsection set based on the second abnormal subsection to be judged;
Obtaining a reference data set of a second reference comparison section set, calculating abnormal difference values of negative data item sets of each abnormal subsection to be judged in the second abnormal subsection to be judged and corresponding data items of the reference data set, and outputting a second abnormal difference value set;
Counting the fluctuation range of each data item difference value in the second abnormal difference value set to obtain an abnormal fluctuation range set;
Comparing the fluctuation ranges of the data items in the abnormal fluctuation range set to obtain a comparison and coincidence degree;
And outputting a primary grade risk judgment result when the comparison coincidence degree reaches a preset variable amplitude coincidence abnormal judgment value.
Preferably, based on the primary level risk judging result, calculating the abnormal difference value of the corresponding data item between the negative data item set of each abnormal subsection to be judged in the second abnormal subsection to be judged and the positive data item set of the positive subsection at the adjacent position, and outputting a third abnormal difference value set;
calculating the ratio between adjacent abnormal differences in the third abnormal difference set to obtain a ratio data set;
Extracting abnormal subsections to be judged corresponding to the ratio data stored with the ratio data larger than a preset standard ratio threshold value in the ratio data set, and outputting a risk abnormal subsection extraction result;
Judging the line fire risk degree of the risk subsection in the risk abnormality subsection extraction result as a second level, and outputting a second level risk judgment result;
And carrying out real-time early warning notification on the first-stage risk judgment result or the second-stage risk judgment result or the third-stage risk judgment result, and outputting a line fire risk real-time early warning notification result.
By adopting the technical scheme, the data contained in the negative subsection in the second abnormal subsection to be judged and the positive subsection in the adjacent joint position are subjected to corresponding difference value calculation, and according to the characteristics of the obtained multiple data information, the judgment of whether the second abnormal subsection to be judged has fire risks or not and the foremost judgment of the first-stage fire risks are realized, the reliability of the fire risk judgment result of the final line is enhanced, multiple verification processing is carried out, and the probability of misjudgment is reduced.
In a second aspect, an artificial intelligence electrical line fire risk real-time early warning system includes:
The line marking unit is used for obtaining a first target line, carrying out abnormal phenomena on all the subsections segmented in the first target line to obtain a negative subsection set, and carrying out normal phenomenon sign marking on all the subsections segmented in the first target line to obtain a positive subsection set;
The first risk judging unit is used for carrying out independent analysis on the number and the positions of the negative sub-section sets to obtain a first abnormal sub-section to be judged, and analyzing the positions of the negative sub-section sets to obtain a first-stage risk judging junction when the number of the negative sub-section sets is smaller than that of the positive sub-section sets;
The second risk judging unit is used for analyzing the position connection degree of the negative sub-section sets to obtain a second risk judging result or a second abnormal sub-section to be judged, and analyzing the comparability of the number of the negative sub-section sets being greater than that of the positive sub-section sets to obtain a first risk judging result;
the third risk judging unit is used for acquiring a second target line, carrying out data analysis on the first reference comparison section segmented in the second target line and the first abnormal subsection to be judged, and outputting a third risk judging result;
the fourth risk judging unit is used for carrying out data analysis on the second reference comparison section set segmented in the second target line and the second abnormal subsection set to be judged and outputting a primary grade risk judging result;
the real-time early warning notification unit is used for carrying out ratio solving processing on the data in the negative subsection set and the positive subsection set according to the primary level risk judgment result, outputting a second level risk judgment result, carrying out real-time early warning notification on the first level risk judgment result or the second level risk judgment result or the third level risk judgment result, and outputting a line fire risk real-time early warning notification result.
Compared with the prior art, the invention has the following characteristics and beneficial effects:
The first target line is segmented into the monitoring section and the reserved section so as to conveniently monitor the fire risk condition of the line, timely real-time early warning notification is carried out in a reserved period before data transmission to the reserved line, the monitoring section is finely segmented into a plurality of sub-section lines, analysis on whether the data of each sub-section is abnormal or not is carried out, and the corresponding sub-section is marked with two different symbols, namely a negative sub-section set and a positive sub-section set, so that the fire risk of the line in different levels is judged according to the number of the sub-sections and the degree of engagement of the two symbols with the position, the accuracy of line fire risk monitoring is improved, and the early warning work efficiency is enhanced.
According to the different relationships between the number and the positions of the negative subsection sets and the positive subsection sets, analysis of various different conditions is performed, further fire risk analysis is performed on the obtained two abnormal subsections to be judged, so that the probability of misjudgment is reduced, the comparison analysis of data difference is performed on the two reference comparison subsection sets and the two abnormal subsections to be judged in the second target line, so that the analyzed data change difference characteristics are convenient to judge the fire risk level of the two abnormal subsections to be judged, the large error caused by the fact that a traditional line fire risk detection technology only performs single analysis on a large number of data resources is avoided, and different levels of early warning is performed on the obtained results under different analysis conditions through the analysis of diversity, so that subsequent maintenance personnel can process timely and pertinently.
Drawings
Fig. 1 is a block diagram of steps of an artificial intelligent electrical line fire risk real-time early warning method mainly embodied in the embodiment.
Fig. 2 is a block diagram of the steps of the S1 substep mainly embodied in the present embodiment.
Fig. 3 is a block diagram of the steps of the S2 substep mainly embodied in this embodiment.
Fig. 4 is a block diagram of the steps of the S3 substep mainly embodied in this embodiment.
Fig. 5 is a block diagram of the steps of the S4 substep mainly embodied in the present embodiment.
Fig. 6 is a block diagram of the steps of the S5 substep mainly embodied in the present embodiment.
Fig. 7 is a block diagram of the steps of the S6 substep mainly embodied in the present embodiment.
Fig. 8 is a block diagram of an artificial intelligent electrical line fire risk real-time early warning system mainly embodied in the present embodiment.
Reference numerals illustrate: 1. a line marking unit; 2. a first decision risk unit; 3. a second decision risk unit; 4. a third decision risk unit; 5. a fourth decision risk unit; 6. and a real-time early warning notification unit.
Detailed Description
The invention is described in further detail below in connection with the following examples.
Referring to fig. 1, an artificial intelligent electrical line fire risk real-time early warning method and system, the method comprises the following steps:
S1, acquiring a first target line, carrying out abnormal phenomena on all subsections segmented in the first target line to obtain a negative subsection set, and carrying out normal phenomenon sign marking on all subsections segmented in the first target line to obtain a positive subsection set.
S2, carrying out independent analysis on the number and the positions of the negative sub-section sets to obtain a first abnormal sub-section to be judged, and analyzing the positions of the negative sub-section sets to obtain a first-stage risk judging result when the number of the negative sub-section sets is smaller than that of the positive sub-section sets.
S3, analyzing the position connection degree of the negative sub-section sets to obtain a second-stage risk judging result or a second abnormal sub-section to be judged, and analyzing the comparability of the number of the negative sub-section sets being greater than that of the positive sub-section sets to obtain a first-stage risk judging result.
S4, acquiring a second target line, carrying out data analysis on the first reference comparison section segmented in the second target line and the first abnormal subsection to be judged, and outputting a third-level risk judgment result.
S5, carrying out data analysis on the second reference comparison section set segmented in the second target line and the second abnormal subsection set to be judged, and outputting a primary judging result of the level risk.
S6, based on the primary level risk judging result, carrying out ratio solving processing on data in the negative subsection set and the positive subsection set, outputting a second level risk judging result, carrying out real-time early warning notification on the first level risk judging result or the second level risk judging result or the third level risk judging result, and outputting a line fire risk real-time early warning notification result.
Specifically, the first target line is segmented, namely the monitoring section and the reserved section, so that early warning notification of timeliness is conveniently carried out on judging risks of a subsequent line fire, the monitoring section is finely segmented into a plurality of equidistant subsections, analysis on whether each subsection is abnormal or not is carried out according to various data of each subsection, marking of two different symbols is carried out on each data item, and corresponding two different symbol marks are carried out on the subsections comprehensively according to the number of the two different symbol marks of the data item, so that subsequent series of analysis processing is carried out on the subsections possibly having abnormal line fire risks in a preliminary mode, the differentiation processing of diversified detectable data information is carried out on unified data information, the accuracy of a data analysis process is greatly improved, differentiation analysis on the obtained positive subsection set and negative subsection set under different conditions of quantity and position is carried out, the severity level degree of the line fire is conveniently known, the early warning notification processing of different levels is carried out, the real-time early warning working efficiency is improved, and the timeliness of the early warning processing is convenient.
Referring to fig. 2, a specific step S1 includes the following sub-steps:
S1001, acquiring a first target line, dividing a monitoring section and a reserved section of the first target line, and carrying out refinement division of at least two equidistant sections on the monitoring section to obtain a first sub-section set.
S1002, collecting a monitoring data set of all subsections in the first subsection set.
S1003, presetting a monitoring judgment section value data set, comparing and analyzing the corresponding section values of the monitoring data set and the monitoring judgment section value data set, and outputting a data judgment normal result and a data judgment abnormal result.
S1004, judging a normal result based on the data, and marking the normal item data with a forward sign to obtain a forward data item set.
S1005, marking negative signs on abnormal item data to obtain a negative data item set based on the data judgment abnormal result.
S1006, calculating the data item number ratio of the positive data item set to the negative data item set, and when the data item number ratio is smaller than or equal to a preset ratio, primarily judging the corresponding sub-section as abnormal, marking the negative sign to obtain a negative sub-section, and outputting the negative sub-section set according to the negative sub-section.
S1007, when the number ratio of the data items is larger than the preset ratio, the corresponding sub-section is primarily judged to be normal, the forward symbol is marked to obtain a forward sub-section, and the forward sub-section set is output according to the forward sub-section.
Specifically, if the length of the first target line is 15 meters, the cut monitoring section is 10 meters, the reserved section is 5 meters, the 10 meters monitoring section is finely cut into 10 subsections lines (the first subsection, the second subsection until the tenth subsection) with equidistant length of 1 meter, namely a first subsection set, monitoring data sets (for example, five items of data including temperature rise data 10 ℃, current data 20A, humidity data 15%rh, transmission rate 100b/s and line loss degree 2%) of all subsections are collected, a monitoring judgment section value data set (for example, five section values respectively corresponding to the five items of data, temperature rise: 0-20 ℃, current: 10A-20A, humidity: 2%rh-10%rh, rate: 120b/s-200b/s and line loss: 0-10%) are preset, and the monitoring data sets and the monitoring judgment section value data sets are subjected to comparative analysis of corresponding section values, and the data judgment result is: the temperature rise data, the current data and the line loss degree are marked by "+" signs to obtain a forward data item set, and the data abnormality judgment result is that: and the humidity data and the transmission rate are marked by using a "-" symbol to obtain negative data item sets, wherein the number of positive data item sets is 3, the number of negative data item sets is 2, the data item number ratio is 3/2, the preset ratio is set to be one half, the data item number ratio is 3/2 and is more than one half, the corresponding second sub-section is primarily judged to be normal, the positive sub-section sets are obtained by marking with a "+" symbol, the corresponding second sub-section is primarily judged to be abnormal if the data item number ratio is 1/4, and the negative sub-section sets are obtained by marking with a "-" symbol.
Referring to fig. 3, a specific step S2 includes the following sub-steps:
s2001, judging the number and the positions of the negative sub-section sets, and when only one negative sub-section set is arranged at the joint of the reserved sections, further analyzing whether the line fire risk exists in the one negative sub-section, and outputting a first abnormal sub-section to be judged.
S2002. when the number of the negative sub-section sets is more than one and less than the number of the positive sub-section sets, and the distribution positions of the negative sub-section sets are located at the junction of the reserved sections and are continuous between each negative sub-section position, the first monitoring condition is output.
And S2003, judging the line fire risk degree of the first monitoring condition as a first level, and outputting a first-level risk judgment result.
Specifically, if the number of negative sub-segment sets is 1 and the negative sub-segment is located at the intersection of the reserved segments (i.e., the tenth sub-segment), and the tenth sub-segment is determined to be the first abnormal sub-segment to be determined, since the first 9 sub-segments are all normal conditions and only the tenth sub-segment is abnormal, in order to avoid erroneous determination, further analysis is required, and when the number of negative sub-segment sets is more than two and less than the number of positive sub-segment sets, and the distribution position of the negative sub-segment sets is located at the intersection of the reserved segments and is continuous between the positions of each negative sub-segment (i.e., the number of negative sub-segment sets is 3, the number of positive sub-segment sets is 7, and the positions of the 3 negative sub-segments are adjacent and continuous: the eighth sub-segment, the ninth sub-segment and the tenth sub-segment, i.e., the first monitored condition), the line fire risk degree of the first monitored condition is one level.
Referring to fig. 4, a specific step S3 includes the following sub-steps:
s3001, outputting a second monitoring condition when the number of the negative sub-section sets is more than one and less than the number of the positive sub-section sets, and the distribution position of the negative sub-section sets does not fall at the junction of the reserved sections.
S3002, calculating the degree of connection of the positions of the negative subsections, and outputting a first degree of connection.
S3003, presetting a connection degree judgment interval value, judging that the line fire risk degree of the second monitoring condition is two-stage according to the first connection degree when the first connection degree falls in the connection degree judgment interval value, and outputting a second-stage risk judgment result.
S3004, when the first degree of engagement does not fall within the degree of engagement judging section value, further analyzing whether the second monitoring condition is at risk of line fire or not, and outputting a second abnormal subsection to be judged.
S3005, when the number of negative subsection sets is greater than that of positive subsection sets, outputting a third monitoring condition, judging the line fire risk degree of the third monitoring condition as a first level, and outputting a first-level risk judgment result.
Specifically, if the number of negative sub-section sets is greater than one and less than the number of positive sub-section sets, and the distribution position of the negative sub-section sets does not fall at the junction of the reserved sections (e.g., the number of negative sub-section sets is 3, the number of positive sub-section sets is 7, and the position of the three negative sub-sections is in the middle or the front of 10 sub-sections, that is, the second monitoring condition), the engagement degree of the position of each negative sub-section is calculated (if the three negative sub-sections are respectively the third sub-section, the fifth sub-section and the eighth sub-section, the third sub-section and the fifth sub-section differ by two sections, and the fifth sub-section differs by 3 sections, then the first engagement degree is calculated to be [ 100/(2+3) ] =20% ], the preset engagement degree determination section value (if 15% -50%), the line risk degree of the second monitoring condition is two, the preset engagement degree determination section value (if 30% -50%), the second line risk degree is further monitored, that is the third sub-section risk condition is more than 6, and the number of sub-sections to be analyzed is the number of positive sub-sections is the abnormal fire risk condition, that is the number of sub-sections is determined to be 6.
Referring to fig. 5, a specific step S4 includes the following sub-steps:
S4001, obtaining a line information base associated with a first target line, and screening a second target line with highest similarity association degree with the first target line from the line information base.
S4002, according to the first sub-section set, the same sub-section is cut on the second target line to obtain a second sub-section set.
S4003, based on the first subsection to be judged, screening a first reference comparison section corresponding to the first subsection to be judged from the second subsection set.
S4004, acquiring a reference data set of a first reference comparison section, calculating an abnormal difference value between a negative data item set of a first subsection to be judged and a corresponding data item of the reference data set, and outputting a first abnormal difference value set.
S4005, evaluating the mutual influence degree among the data item difference values of the first abnormal difference value set, and outputting a first influence degree evaluation result.
S4006, presetting an influence degree critical judgment value, and outputting a third-stage risk judgment result when the numerical value in the first influence degree evaluation result is greater than or equal to the influence degree critical judgment value.
Specifically, if the line length of the second target line is also 15 meters, the second target line is segmented into a plurality of sub-sections corresponding to the first target line, a second sub-section set (also 10) is obtained, a first reference comparison section corresponding to the first to-be-judged abnormal sub-section is screened out from the second sub-section set (if the first to-be-judged abnormal sub-section is the tenth sub-section and the twelfth sub-section is the twenty-second sub-section), the first reference comparison section is the twenty-second sub-section, the first reference comparison section is the reference data set of the first reference comparison section (if the temperature rise data is 15 ℃, the current data is 20A, the humidity data is 10% rh, the transmission rate is 160b/s and the line loss degree is 8%), the first to-be-judged abnormal difference value is calculated between the first to-be-abnormal data item set and the reference data set (if the analysis knows that the abnormal data item is the humidity data is 15% rh and the transmission rate is 100b/s, the calculated first abnormal difference value is 5% rhb/s), and if the first to-abnormal data item is 60% and the first to-abnormal data item is 60%, the first to-abnormal data item is 5% and the first to-abnormal data item is 8%, the first to-abnormal data item is calculated to be the abnormal difference value, and the first to be the abnormal data item is 8% is calculated, and the abnormal data item is calculated, if the first to be the abnormal data item is 5% has a high to be 8% and the abnormal rate, and the abnormal data is 8.
Referring to fig. 6, a specific step S5 includes the following sub-steps:
s5001, based on the second abnormal subsection to be judged, screening a second reference comparison section set corresponding to the position of the second abnormal subsection to be judged from the second subsection set.
S5002, acquiring a reference data set of a second reference comparison section set, calculating abnormal difference values of negative data item sets of each abnormal subsection to be judged in the second abnormal subsection to be judged and corresponding data items of the reference data set, and outputting a second abnormal difference value set.
S5003, counting the fluctuation range of the difference values of each data item in the second abnormal difference value set to obtain an abnormal fluctuation range set.
S5004, comparing the fluctuation ranges of the data items in the abnormal fluctuation range set to obtain comparison weight.
S5005, outputting a primary judging result of the level risk when the comparison coincidence degree reaches a preset amplitude-variable coincidence abnormal judging value.
Specifically, if a second reference comparison section set (corresponding to the plurality of second to-be-judged abnormal subsections) corresponding to the position of the second to-be-judged abnormal subsections is selected from the second subsection set, the abnormal difference value is calculated between the negative data item set of each to-be-judged abnormal subsection in the second to-be-judged abnormal subsections and the corresponding data item of the reference data set in the same processing step (if the second to-be-judged abnormal subsections set is the third subsection, the fifth subsection and the eighth subsection respectively, the acquired second abnormal difference value set is that the humidity data difference value is 5% rh, 8% rh and 10% rh respectively, the transmission rate difference value is 40b/s, 50b/s and 60b/s respectively), the fluctuation range of the difference values of all the data items is 3% rh, 2% rh, and the rates 10b/s and 10b/s according to the positions, the fluctuation range of all the data items in the abnormal fluctuation range set is compared to obtain a comparison weight (if the acquired second abnormal difference value set is 5% rh, 8% rh and 10% rh is the preset to the corresponding to the amplitude of the abnormal value of the abnormal comparison table), and if the acquired second abnormal difference value set is the abnormal difference value is the preliminary fire risk of the comparison weight.
Referring to fig. 7, a specific step S6 includes the following sub-steps:
s6001, based on the primary level risk judging result, calculating abnormal difference values of corresponding data items between the negative data item set of each abnormal subsection to be judged in the second abnormal subsection to be judged and the positive data item set of the positive subsection at the adjacent position, and outputting a third abnormal difference value set.
S6002, calculating the ratio between adjacent abnormal differences in the third abnormal difference set to obtain a ratio data set.
S6003, extracting abnormal subsections to be judged corresponding to the ratio data with the ratio data larger than a preset standard ratio threshold value in the ratio data set, and outputting risk abnormal subsections extraction results.
S6004, judging the line fire risk degree of the risk subsection in the risk abnormality subsection extraction result as a second grade, and outputting a second grade risk judgment result.
S6005, carrying out real-time early warning notification on the first-stage risk judgment result or the second-stage risk judgment result or the third-stage risk judgment result, and outputting a line fire risk real-time early warning notification result.
Specifically, if the initial judging result of the level risk is obtained, continuing to analyze and judge whether the fire risk and the fire risk level of the second abnormal subsection to be judged are in the next step, and calculating the abnormal difference value of the corresponding data item between the negative data item set of each abnormal subsection to be judged in the second abnormal subsection to be judged and the positive data item set of the positive subsection at the adjacent position (if the thirteenth subsection, the fifteenth subsection and the eighteenth subsection are negative subsections, the positive subsection at the adjacent position can be extracted: the third abnormal difference value set between the three positive sub-sections and the corresponding item data of the abnormal item data of the three negative sub-sections is a humidity difference value of 4%rh, 6%rh and 7%rh, the velocity difference value of 40b/s, 50b/s and 80b/s, the ratio of adjacent abnormal difference values in the third abnormal difference value set is calculated to obtain a ratio data set (humidity difference value ratio of 2/3, 6/7, velocity difference value ratio of 4/5 and 5/8), a standard ratio threshold value is preset (if the ratio is 4/5), the second abnormal sub-section to be judged corresponding to the humidity difference value ratio of 6/7 and the velocity difference value ratio of 5/8 is judged to be a line fire risk degree secondary, and the analyzed and judged risk judgment results under the various different conditions are all notified to be transmitted, so that effective targeted processing is carried out by subsequent maintenance personnel, and the high-efficiency early warning of the whole line fire risk real-time early warning and processing work is improved.
An artificial intelligent electric circuit fire risk real-time early warning system comprises a circuit marking unit 1, a first judging risk unit 2, a second judging risk unit 3, a third judging risk unit 4, a fourth judging risk unit 5 and a real-time early warning notification unit 6, wherein referring to fig. 8, a first target circuit is obtained through the circuit marking unit 1, abnormal phenomena are carried out on all sub-sections segmented in the first target circuit to obtain a negative sub-section set, and normal phenomena are carried out on all sub-sections segmented in the first target circuit to obtain a positive sub-section set; the number and the positions of the negative sub-section sets are subjected to independent analysis through a first judging risk unit 2 to obtain a first abnormal sub-section to be judged, and when the number of the negative sub-section sets is smaller than that of the positive sub-section sets, the positions of the negative sub-section sets are analyzed to obtain a first-stage risk judging result; analyzing the position connection degree of the negative sub-section sets through a second judging risk unit 3 to obtain a second-stage risk judging result or a second abnormal sub-section to be judged, and analyzing the comparability of the number of the negative sub-section sets being greater than that of the positive sub-section sets to obtain a first-stage risk judging result; acquiring a second target line through a third judging risk unit 4, carrying out data analysis on a first reference comparison section segmented in the second target line and a first abnormal subsection to be judged, and outputting a third-level risk judging result; the fourth judging risk unit 5 is used for carrying out data analysis on the second reference comparison section set segmented in the second target line and the second abnormal subsection set to be judged, and outputting a primary judging result of the level risk; and carrying out ratio solving processing on data in the negative subsection set and the positive subsection set according to the primary level risk judging result by the real-time early warning informing unit 6, outputting a second level risk judging result, carrying out real-time early warning informing on the first level risk judging result or the second level risk judging result or the third level risk judging result, and outputting a line fire risk real-time early warning informing result.
The above embodiments are not intended to limit the scope of the present application, so: all equivalent changes in structure, shape and principle of the application should be covered in the scope of protection of the application.
Claims (8)
1. The real-time early warning method for the fire risk of the artificial intelligent electric circuit is characterized by comprising the following steps of:
Acquiring a first target line, performing anomaly on all subsections segmented in the first target line to obtain a negative subsection set, and performing normal sign marking on all subsections segmented in the first target line to obtain a positive subsection set;
Performing independent analysis on the number of the negative sub-section sets and the positions of the negative sub-section sets to obtain a first abnormal sub-section to be judged, and analyzing the positions of the negative sub-section sets to obtain a first-stage risk judgment result when the number of the negative sub-section sets is smaller than the number of the positive sub-section sets;
analyzing the position connection degree of the negative sub-section set to obtain a second-stage risk judging result or a second abnormal sub-section to be judged, and analyzing the comparability of the number of the negative sub-section sets being greater than that of the positive sub-section sets to obtain a first-stage risk judging result;
Acquiring a second target line, carrying out data analysis on a first reference comparison section segmented in the second target line and a first abnormal subsection to be judged, and outputting a third-level risk judgment result;
Performing data analysis on the second reference comparison section set segmented in the second target line and the second abnormal subsection set to be judged, and outputting a primary level risk judgment result;
And based on the primary level risk judging result, carrying out ratio solving processing on the data in the negative subsection set and the positive subsection set, outputting a second level risk judging result, carrying out real-time early warning notification on the first level risk judging result or the second level risk judging result or the third level risk judging result, and outputting a line fire real-time early warning notification result.
2. The real-time early warning method for fire risk of an artificial intelligent electric circuit according to claim 1, wherein the method comprises the steps of obtaining a first target circuit, obtaining a negative sub-section set by performing an anomaly on all sub-sections cut in the first target circuit, and obtaining a positive sub-section set by performing a sign marking of a normal phenomenon on all sub-sections cut in the first target circuit, and is specifically as follows:
Acquiring a first target line, dividing a monitoring section and a reserved section of the first target line, and carrying out refinement division of at least two equidistant sections on the monitoring section to obtain a first sub-section set;
Collecting a monitoring dataset of all subsections in the first subsection set;
Presetting a monitoring judgment section value data set, carrying out comparative analysis on corresponding section values on the monitoring data set and the monitoring judgment section value data set, and outputting a data judgment normal result and a data judgment abnormal result;
based on the data, judging a normal result, and marking a forward symbol on normal item data to obtain a forward data item set;
Based on the data judging abnormal result, marking the abnormal item data with negative sign to obtain a negative data item set;
Calculating the data item number ratio of the positive data item set to the negative data item set, and preliminarily judging the corresponding sub-section as abnormal when the data item number ratio is smaller than or equal to a preset ratio, marking negative signs to obtain a negative sub-section, and outputting the negative sub-section set according to the negative sub-section;
When the number ratio of the data items is larger than the preset ratio, the corresponding sub-section is primarily judged to be normal, the forward symbol is marked to obtain a forward sub-section, and the forward sub-section set is output according to the forward sub-section.
3. The real-time early warning method for fire risk of an artificial intelligence electric circuit according to claim 2, wherein the step of analyzing the number of the negative sub-section sets and the positions thereof to obtain the first abnormal sub-section to be judged, and analyzing the positions of the negative sub-section sets to obtain the first-stage risk judging result when the number of the negative sub-section sets is smaller than the number of the positive sub-section sets is specifically as follows:
Judging the number and the positions of the negative sub-section sets, and when only one negative sub-section set is arranged and the negative sub-section is positioned at the junction of the reserved sections, further analyzing whether the risk of the line fire exists in the negative sub-section or not, and outputting a first abnormal sub-section to be judged;
Outputting a first monitoring condition when the number of the negative sub-section sets is more than one and less than the number of the positive sub-section sets, and the distribution positions of the negative sub-section sets are located at the junction of the reserved sections and are continuous among the positions of each negative sub-section;
Judging the line fire risk degree of the first monitoring condition as a first level, and outputting a first-level risk judgment result.
4. The real-time early warning method for fire risk of an artificial intelligence electric circuit according to claim 2, wherein the step of analyzing the degree of engagement of the positions of the negative sub-section sets to obtain a second-stage risk determination result or a second abnormal sub-section to be determined, and analyzing the comparability of the number of the negative sub-section sets being greater than that of the positive sub-section sets to obtain a first-stage risk determination result comprises the following steps:
Outputting a second monitoring condition when the number of the negative sub-section sets is more than one and less than the number of the positive sub-section sets and the distribution position of the negative sub-section sets does not fall at the junction of the reserved sections;
Calculating the degree of engagement of the positions of each negative sub-section, and outputting a first degree of engagement;
Presetting a connection degree judging section value, judging the line fire risk degree of a second monitoring condition as a second level according to the first connection degree when the first connection degree falls in the connection degree judging section value, and outputting a second level risk judging result;
When the first degree of engagement does not fall within the degree of engagement judging section value, further analyzing whether the second monitoring condition is in existence of the fire risk of the line, and outputting a second abnormal subsection to be judged;
When the number of the negative subsection sets is greater than that of the positive subsection sets, outputting a third monitoring condition, judging the line fire risk degree of the third monitoring condition as a first level, and outputting a first-level risk judgment result.
5. The real-time early warning method for fire risk of artificial intelligence electric circuit according to claim 3, wherein the step of obtaining a second target circuit, comparing the first reference comparison section cut in the second target circuit with the first abnormal subsection to be judged, and outputting a third-stage risk judgment result comprises the following steps:
acquiring a line information base associated with a first target line, and screening a second target line with highest similarity association degree with the first target line from the line information base;
According to the first sub-section set, the same sub-section of the second target line is segmented to obtain a second sub-section set;
Screening a first reference comparison section corresponding to the first abnormal subsection to be judged from the second subsection set based on the first abnormal subsection to be judged;
acquiring a reference data set of a first reference comparison section, calculating an abnormal difference value between a negative data item set of the first abnormal subsection to be judged and a corresponding data item of the reference data set, and outputting a first abnormal difference value set;
evaluating the mutual influence degree of the data item differences in the first abnormal difference set, and outputting a first influence degree evaluation result;
Presetting an influence degree critical judgment value, and outputting a third-stage risk judgment result when the numerical value in the first influence degree evaluation result is larger than or equal to the influence degree critical judgment value.
6. The method for real-time early warning of fire risk of an artificial intelligent electrical circuit according to claim 4, wherein the step of analyzing data between the second reference comparison section set and the second abnormal subsection set to be judged, and outputting the primary judgment result of the level risk is specifically as follows:
Screening a second reference comparison section set corresponding to the position of the second abnormal subsection set to be judged from the second subsection set based on the second abnormal subsection to be judged;
Obtaining a reference data set of a second reference comparison section set, calculating abnormal difference values of negative data item sets of each abnormal subsection to be judged in the second abnormal subsection to be judged and corresponding data items of the reference data set, and outputting a second abnormal difference value set;
Counting the fluctuation range of each data item difference value in the second abnormal difference value set to obtain an abnormal fluctuation range set;
Comparing the fluctuation ranges of the data items in the abnormal fluctuation range set to obtain a comparison and coincidence degree;
And outputting a primary grade risk judgment result when the comparison coincidence degree reaches a preset variable amplitude coincidence abnormal judgment value.
7. The method for real-time early warning of fire risk of an artificial intelligent electric circuit according to claim 6, wherein based on the primary judgment result of the level risk, the data in the negative subsection set and the positive subsection set are subjected to ratio processing, the second-level risk judgment result is output, the first-level risk judgment result or the second-level risk judgment result or the third-level risk judgment result is subjected to real-time early warning notification, and the real-time early warning notification result of the fire risk of the circuit is output, specifically comprising the steps of:
based on the primary level risk judging result, calculating the abnormal difference value of the corresponding data item between the negative data item set of each abnormal subsection to be judged in the second abnormal subsection to be judged and the positive data item set of the positive subsection at the adjacent position, and outputting a third abnormal difference value set;
calculating the ratio between adjacent abnormal differences in the third abnormal difference set to obtain a ratio data set;
Extracting abnormal subsections to be judged corresponding to the ratio data stored with the ratio data larger than a preset standard ratio threshold value in the ratio data set, and outputting a risk abnormal subsection extraction result;
Judging the line fire risk degree of the risk subsection in the risk abnormality subsection extraction result as a second level, and outputting a second level risk judgment result;
And carrying out real-time early warning notification on the first-stage risk judgment result or the second-stage risk judgment result or the third-stage risk judgment result, and outputting a line fire risk real-time early warning notification result.
8. A system for real-time early warning of fire risk of an artificial intelligent electrical circuit, which is characterized in that the system is used for realizing the real-time early warning method of fire risk of an artificial intelligent electrical circuit according to any one of claims 1-7, and comprises the following steps:
The line marking unit is used for obtaining a first target line, carrying out abnormal phenomena on all the subsections segmented in the first target line to obtain a negative subsection set, and carrying out normal phenomenon sign marking on all the subsections segmented in the first target line to obtain a positive subsection set;
The first risk judging unit is used for carrying out independent analysis on the number and the positions of the negative sub-section sets to obtain a first abnormal sub-section to be judged, and analyzing the positions of the negative sub-section sets to obtain a first-stage risk judging junction when the number of the negative sub-section sets is smaller than that of the positive sub-section sets;
The second risk judging unit is used for analyzing the position connection degree of the negative sub-section sets to obtain a second risk judging result or a second abnormal sub-section to be judged, and analyzing the comparability of the number of the negative sub-section sets being greater than that of the positive sub-section sets to obtain a first risk judging result;
the third risk judging unit is used for acquiring a second target line, carrying out data analysis on the first reference comparison section segmented in the second target line and the first abnormal subsection to be judged, and outputting a third risk judging result;
the fourth risk judging unit is used for carrying out data analysis on the second reference comparison section set segmented in the second target line and the second abnormal subsection set to be judged and outputting a primary grade risk judging result;
the real-time early warning notification unit is used for carrying out ratio solving processing on the data in the negative subsection set and the positive subsection set according to the primary level risk judgment result, outputting a second level risk judgment result, carrying out real-time early warning notification on the first level risk judgment result or the second level risk judgment result or the third level risk judgment result, and outputting a line fire risk real-time early warning notification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410455501.2A CN118053280B (en) | 2024-04-16 | 2024-04-16 | Real-time early warning method and system for fire risk of artificial intelligent electric circuit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410455501.2A CN118053280B (en) | 2024-04-16 | 2024-04-16 | Real-time early warning method and system for fire risk of artificial intelligent electric circuit |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118053280A true CN118053280A (en) | 2024-05-17 |
| CN118053280B CN118053280B (en) | 2024-07-19 |
Family
ID=91046995
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410455501.2A Active CN118053280B (en) | 2024-04-16 | 2024-04-16 | Real-time early warning method and system for fire risk of artificial intelligent electric circuit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118053280B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101717775B1 (en) * | 2016-10-17 | 2017-03-17 | 이성원 | System for Analysising and Predicting Electric Fire using Intelligent Pre-signal Analysis |
| CN107564231A (en) * | 2017-09-15 | 2018-01-09 | 山东建筑大学 | Building fire early warning and fire disaster situation assessment system and method based on Internet of Things |
| CN111145517A (en) * | 2020-01-03 | 2020-05-12 | 上海枫昱能源科技有限公司 | Artificial intelligence electric line fire risk real-time early warning method and system |
| CN117292528A (en) * | 2023-09-27 | 2023-12-26 | 华筑实业(杭州)股份有限公司 | Real-time early warning method and system for fire risk of electric circuit |
| CN117523808A (en) * | 2024-01-04 | 2024-02-06 | 珠海派诺科技股份有限公司 | Electrical fire early warning system and method capable of being monitored in real time based on Internet of things |
-
2024
- 2024-04-16 CN CN202410455501.2A patent/CN118053280B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101717775B1 (en) * | 2016-10-17 | 2017-03-17 | 이성원 | System for Analysising and Predicting Electric Fire using Intelligent Pre-signal Analysis |
| CN107564231A (en) * | 2017-09-15 | 2018-01-09 | 山东建筑大学 | Building fire early warning and fire disaster situation assessment system and method based on Internet of Things |
| CN111145517A (en) * | 2020-01-03 | 2020-05-12 | 上海枫昱能源科技有限公司 | Artificial intelligence electric line fire risk real-time early warning method and system |
| CN117292528A (en) * | 2023-09-27 | 2023-12-26 | 华筑实业(杭州)股份有限公司 | Real-time early warning method and system for fire risk of electric circuit |
| CN117523808A (en) * | 2024-01-04 | 2024-02-06 | 珠海派诺科技股份有限公司 | Electrical fire early warning system and method capable of being monitored in real time based on Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118053280B (en) | 2024-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583680B (en) | Power stealing identification method based on support vector machine | |
| JP6207078B2 (en) | Monitoring device, monitoring method and program | |
| CN114201374B (en) | Operation and maintenance time sequence data anomaly detection method and system based on hybrid machine learning | |
| CN112327100B (en) | Power failure detection method and system based on Internet of things | |
| CN104994334A (en) | Automatic substation monitoring method based on real-time video | |
| CN105117512B (en) | The evaluation method and device of transformer early warning value | |
| CN110580492A (en) | Track circuit fault precursor discovery method based on small fluctuation detection | |
| CN115865649A (en) | Intelligent operation and maintenance management control method, system and storage medium | |
| CN111157850A (en) | Mean value clustering-based power grid line fault identification method | |
| CN113536440A (en) | Data processing method based on BIM operation and maintenance management system | |
| CN118053280B (en) | Real-time early warning method and system for fire risk of artificial intelligent electric circuit | |
| CN113128707A (en) | Situation risk assessment method for distribution automation terminal | |
| CN115410342A (en) | Landslide disaster intelligent early warning method based on crack meter real-time monitoring | |
| CN109298285A (en) | A kind of identification of distribution network cable initial failure and early warning system and method based on transient disturbance | |
| CN113589098A (en) | Power grid fault prediction and diagnosis method based on big data drive | |
| CN108731731A (en) | A kind of lighning proof type safety supervision system and lighning proof type safety supervision method | |
| CN117556318A (en) | Early warning method and device of cable network identification system | |
| CN114167837B (en) | Intelligent fault diagnosis method and system for railway signal system | |
| CN110674193A (en) | Intelligent substation relay protection fault information modeling method | |
| CN115796832A (en) | Comprehensive evaluation method for health state of power transformation equipment based on multidimensional parameters | |
| CN112434955A (en) | Distribution network line operation risk sensing method based on multi-data fusion | |
| CN103217621A (en) | Power quality monitoring method and power quality monitoring system | |
| CN117935519B (en) | Gas detection alarm system | |
| CN117374976B (en) | Electrical safety management system based on automatic line fault identification | |
| CN209690438U (en) | A kind of identification of distribution network cable initial failure and early warning system based on transient disturbance |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |